FRAGSTATS SPATIAL PATTERN ANALYSIS PROGRAM FOR QUANTIFYING LANDSCAPE STRUCTURE. Version 2.0 KEVIN MCGARIGAL 1

Transcription

1 FRAGSTATS SPATIAL PATTERN ANALYSIS PROGRAM FOR QUANTIFYING LANDSCAPE STRUCTURE Version 2.0 by KEVIN MCGARIGAL 1 Forest Science Department, Oregon State University, Corvallis, OR (303) BARBARA J. MARKS Forest Science Department, Oregon State University, Corvallis, OR (503) March, Present address: P.O. Box 606, Dolores, Colorado [(303) ]

2 PREFACE and 2.0 UPGRADE INFORMATION As the authors of FRAGSTATS, we are VERY concerned about the potential for misuse of this program. Like most tools, FRAGSTATS is only as "good" as the user. FRAGSTATS crunches out a lot of numbers about the input landscape. These numbers can easily become "golden" in the hands of uninformed users. Unfortunately, the "garbage in-garbage out" axiom applies here. We have done our best in the documentation to stress the importance of defining landscape, patch, matrix, and landscape context at a scale and in a manner that is relevant and meaningful to the phenomenon under consideration. Moreover, we have stressed the importance of understanding the exact meaning of each metric before it is used. These and other important considerations in any landscape structural analysis are discussed in the documentation. We strongly urge you to read the entire documentation, especially the section on Key Concepts and Terminology, before ever running FRAGSTATS. We wish to remind users that we are not in the commercial software marketing business. We are scientists who recognized the need for a tool like FRAGSTATS to assist us in our research on landscape ecological issues. Therefore, we do not wish to spend a great deal of time consulting on trivial matters concerning the use of FRAGSTATS. However, we do recognize an obligation to provide some level of information support. Of course, we welcome and encourage your criticisms and suggestions about the program at all times. We will welcome questions about how to run FRAGSTATS or interpret the output only after the user has read the entire documentation. This is only fair and will eliminate many trivial questions. Finally, we are always interested in learning about how others have applied FRAGSTATS in ecological investigations and management applications. Therefore, we encourage you to contact us and describe your application after using FRAGSTATS. This release of FRAGSTATS (version 2.0) differs from the previous version in only minor ways. Several "bugs" have been corrected. The most important change is the added option to treat a specified proportion of the landscape boundary and background edge (instead of just all or none) as true edge in the edge metrics (bound_wght option). This fraction also is used as the edge contrast weight for landscape boundary and background edge segments in the calculation of edge contrast metrics. In addition, the convention for naming the output file containing patch ID s in the raster version has been modified to comply with DOS requirements on a PC (id_image option). Similarly, the output file name extensions in the PC raster version have been shortened and renamed to comply with DOS requirements and to avoid conflicts with ERDAS conventions (out_file). The Nearest-neighbor algorithm has been modified slightly to compute actual edge-to-edge distance (previous version used cell midpoints rather than edge). We hope that FRAGSTATS is of great assistance in your work and we look forward to hearing about your applications.

3 ABSTRACT Landscape ecology involves the study of landscape patterns, the interactions among patches within a landscape mosaic, and how these patterns and interactions change over time. In addition, landscape ecology involves the application of these principles in the formulation and solving of real-world problems. Landscape ecology is largely founded on the notion that the patterning of landscape elements (patches) strongly influences ecological characteristics, including vertebrate populations. The ability to quantify landscape structure is prerequisite to the study of landscape function and change. For this reason, much emphasis has been placed on developing methods to quantify landscape structure. Most of the efforts to date have been tailored to meet the needs of specific research objectives and have employed user-generated computer programs to perform the analyses. Such user-generated programs allow for the inclusion of customized analytical methods and easy linkages to other programs such as spatial simulation models, yet generally lack the advanced graphics capabilities of commercially available GIS. Most of these user-generated programs are limited to a particular hardware environment, are embedded within a larger software package designed to accomplish a specific research objective, do not offer a broad array of landscape metrics, and are designed to analyze raster images only. This report describes a program called FRAGSTATS that we developed to quantify landscape structure. FRAGSTATS offers a comprehensive choice of landscape metrics and was designed to be as versatile as possible. Moreover, the program is almost completely automated and thus requires little technical training. Two separate versions of FRAGSTATS exist; one for vector images and one for raster images. The vector version is an Arc/Info AML that accepts Arc/Info polygon coverages. The raster version is a C program that accepts ASCII image files, 8 or 16 bit binary image files, Arc/Info SVF files, Erdas image files, and IDRISI image files. Both versions of FRAGSTATS generate the same array of metrics, including a variety of area metrics, patch density, size and variability metrics, edge metrics, shape metrics, core area metrics, diversity metrics, and contagion and interspersion metrics. The raster version also computes several nearestneighbor metrics. In this report, each metric calculated by FRAGSTATS is described in terms of its ecological application and limitations. Example landscapes are included and a discussion is provided of each metric as it relates to the sample landscapes. In addition, several important concepts and definitions critical to the assessment of landscape structure are discussed. The appendices include a complete list of algorithms, the units and ranges of each metric, examples of the FRAGSTATS output files, and a users guide that describes in detail how to install and run FRAGSTATS.

4 ACKNOWLEDGEMENTS and DISCLAIMER Many individuals provided valuable feedback during the development and many revisions of the software, including Steve Garman, Eric Gustafson, Jeff Nighbert, Tom Moore, Catherine Rogers, and David Wallin. We are especially grateful to Catherine Rogers and Eric Gustafson for their comprehensive and detailed testing of the program and their many useful suggestions. We thank Jim Kiser for conducting the photogrammetry and digitization work on the landscapes used in the documentation examples. Initial funding for this project was provided through the Coastal Oregon Productivity Enhancement (COPE) program; COPE is a cooperative research and technology transfer effort among Oregon State University, USDA Forest Service, USDI Bureau of Land Management, other state and federal agencies, forest industry, county governments, and resource protection organizations. Subsequent funding for the completion of this documentation was provided by the USDI Bureau of Land Management Cooperative Research Unit and the USDA Forest Service, Pacific Northwest Research Station, Corvallis, Oregon. This software is in the public domain, and the recipient may not assert any proprietary rights thereto nor represent it to anyone as other than an Oregon State University-produced program. FRAGSTATS is provided "as-is" without warranty of any kind, including, but not limited to, the implied warranties of merchantability and fitness for a particular purpose. The user assumes all responsibility for the accuracy and suitability of this program for a specific application. In no event will the authors or Oregon State University be liable for any damages, including lost profits, lost savings, or other incidental or consequential damages arising from the use of or the inability to use this program.

5 TABLE OF CONTENTS Page INTRODUCTION 1 CONCEPTS AND DEFINITIONS 3 FRAGSTATS OVERVIEW 14 FRAGSTATS METRICS 23 General Considerations 23 Area Metrics 25 Patch Density, Size and Variability Metrics 29 Edge Metrics 33 Shape Metrics 39 Core Area Metrics 43 Nearest-Neighbor Metrics 50 Diversity Metrics 54 Contagion and Interspersion Metrics 57 LITERATURE CITED 60 APPENDICES 68 Appendix A. Example of the FRAGSTATS output file formatted exclusively for display purposes (i.e., "basename".full). Each run of FRAGSTATS on a landscape produces an output file like this one. The results reported here correspond to the landscape displayed in figure 6 (landscape B). The results obtained using the vector and raster versions of FRAGSTATS are included separately; note the differences in indices involving edge lengths and patch perimeters. Appendix B. FRAGSTATS user guidelines. Appendix C. Definition and description of FRAGSTATS metrics.

6 LIST OF FIGURES Figure Page 1. Multi-scale view of "landscape" from an organism-centered perspective. Because the eagle, cardinal, and butterfly perceive their environments differently and at different scales, what constitutes a single habitat patch for the eagle may constitute an entire landscape or patch-mosaic for the cardinal, and a single habitat patch for the cardinal may comprise an entire landscape for the butterfly that perceives patches on an even finer scale Alternative image formats accepted in the vector version of FRAGSTATS. Landscape boundary, background, and border are defined in the text Alternative image formats accepted in the raster version of FRAGSTATS. Landscape boundary, background, and border are defined in the text Example of FRAGSTATS patch indices for 3 sample patches drawn from a sample landscape. See text and Appendix C for a description and definition of each metric. Indices with a * were computed using the raster version of FRAGSTATS Example of FRAGSTATS class indices for the mixed, large sawtimber (MLS) patch type in 3 sample landscapes. See text and Appendix C for a description and definition of each metric. Indices with a * were computed using the raster version of FRAGSTATS Example of FRAGSTATS landscape indices for 3 sample landscapes. See text and Appendix C for a description and definition of each metric. Indices with a * were computed using the raster version of FRAGSTATS. 22

7 LIST OF TABLES Table Page 1. Metrics computed in FRAGSTATS, grouped by subject area. See Appendix C for a mathematical definition of each metric. 26

8 INTRODUCTION Growing concerns over the loss of biodiversity has spurred land managers to seek better ways of managing landscapes at a variety of spatial and temporal scales. A number of developments have made possible the ability to analyze and manage entire landscapes to meet multi-resource objectives. The developing field of landscape ecology has provided a strong conceptual and theoretical basis for understanding landscape structure, function, and change (Forman and Godron 1986, Urban et al. 1987, Turner 1989). Growing evidence that habitat fragmentation is detrimental to many species and may contribute substantially to the loss of regional and global biodiversity (Saunders et al. 1991, Harris 1984) has provided empirical justification for the need to manage entire landscapes, not just the components. The development of GIS technology, in particular, has made a variety of analytical tools available for analyzing and managing landscapes. In response to this growing theoretical and empirical support and technical capabilities, public land management agencies have begun to recognize the need to manage natural resources at the landscape scale. A good example of these changes is in the field of wildlife science. Wildlife ecologists often have assumed that the most important ecological processes affecting wildlife populations and communities operate at local spatial scales (Dunning et al. 1992). Vertebrate species richness and abundance, for example, often are considered to be functions of variation in local resource availability, vegetation composition and structure, and the size of the habitat patch (MacArthur and MacArthur 1961, Willson 1974, Cody 1985). Correspondingly, most wildlife research and management activities have focussed on the within-patch scale, typically small plots or forest stands. Wildlife ecologists have become increasingly aware, however, that habitat variation and its affects on ecological processes and vertebrate populations occurs at a wide range of spatial scales (Wiens 1989a,b). In particular, there has been increasing awareness of the potential importance of coarse-scale habitat patterns to wildlife populations, and there has been a corresponding surge in landscape ecological investigations that examine vertebrate distributions and population dynamics over broader spatial scales (e.g., McGarigal and McComb 1994). The recent attention on metapopulation theory (Gilpin and Hanski 1991) and the proliferation of mathematical models on dispersal and spatially distributed populations (Kareiva 1990) are testimony to these changes. Moreover, recent conservation efforts for the northern spotted owl (Strix occidentalis caurina) demonstrate the willingness and ability of public land management agencies to analyze and manage wildlife populations at the landscape scale (Lamberson et al. 1992, Murphy and Noon 1992, Thomas et al. 1990). The emergence of landscape ecology to the forefront of ecology is testimony to the growing recognition that ecological processes affect and are affected by the dynamic interaction among ecosystems. This surge in interest in landscape ecology also has become manifest in a wave of recent efforts to incorporate a landscape perspective into policies and guidelines for managing public lands. Landscape ecology embodies a way of thinking that many see as very useful for organizing land management approaches.

9 Specifically, landscape ecology focusses on 3 characteristics of the landscape (Forman and Godron, 1986): "(1) Structure, the spatial relationships among the distinctive ecosystems or "elements" present--more specifically, the distribution of energy, materials, and species in relation to the sizes, shapes, numbers, kinds, and configurations of the ecosystems. (2) Function, the interactions among the spatial elements, that is, the flows of energy, materials, and species among the component ecosystems. (3) Change, the alteration in the structure and function of the ecological mosaic over time." Thus, landscape ecology involves the study of landscape patterns, the interactions among patches within a landscape mosaic, and how these patterns and interactions change over time. In addition, landscape ecology involves the application of these principles in the formulation and solving of real-world problems. Landscape ecology considers the development and dynamics of spatial heterogeneity and its affects on ecological processes, and the management of spatial heterogeneity (Risser et al. 1984). Landscape ecology is largely founded on the notion that the patterning of landscape elements (patches) strongly influences ecological characteristics, including vertebrate populations. The ability to quantify landscape structure is prerequisite to the study of landscape function and change. For this reason, much emphasis has been placed on developing methods to quantify landscape structure (e.g., O Neill et al. 1988, Li 1990, Turner 1990a, Turner and Gardner 1991). Most of the efforts to date have been tailored to meet the needs of specific research objectives and have employed user-generated computer programs to perform the analyses. Such user-generated programs allow for the inclusion of customized analytical methods and easy linkages to other programs such as spatial simulation models, yet generally lack the advanced graphics capabilities of commercially available GIS (Turner 1990a). Most of these user-generated programs are limited to a particular hardware environment or are embedded within a larger software package designed to accomplish a specific research objective (e.g., to model fire disturbance regimes). Of the available software programs that we are aware of, none offer a broad array of landscape metrics and all are designed to analyze raster images only. This report describes a program called FRAGSTATS that we developed to quantify landscape structure. FRAGSTATS offers a comprehensive choice of landscape metrics and was designed to be as versatile as possible. Moreover, the program is almost completely automated and thus requires little technical training. Two separate versions of 2

10 FRAGSTATS exist; one for vector images and one for raster images. The vector version is an Arc/Info AML that accepts Arc/Info polygon coverages. The raster version is a C program that accepts ASCII image files, 8 or 16 bit binary image files, Arc/Info SVF files, Erdas image files, and IDRISI image files. Both versions of FRAGSTATS generate the same array of metrics, although a few additional metrics are computed in the raster version. In this report, each metric calculated by FRAGSTATS is described in terms of its ecological application and limitations. Example landscapes are included and a discussion is provided of each metric as it relates to the sample landscapes. In addition, several important concepts and definitions critical to the assessment of landscape structure are discussed. The appendices include a complete list of algorithms, the units and ranges of each metric, examples of the FRAGSTATS output files, and a users guide that describes in detail how to install and run FRAGSTATS. 3 CONCEPTS AND DEFINITIONS It is beyond the scope and purpose of this document to provide a glossary of terms and a comprehensive discussion of the many concepts embodied in landscape ecology. Instead, a few key terms and concepts essential to the use of FRAGSTATS and the measurement of spatial heterogeneity are defined and discussed; a thorough understanding of these concepts is prerequisite to the effective use of FRAGSTATS. Landscape.--What is a "landscape"? Surprisingly, there are many different interpretations of this well-used term. The disparity in definitions makes it difficult to communicate clearly, and even more difficult to establish consistent management policies. Definitions of landscape invariably include an area of land containing a mosaic of patches or landscape elements. Forman and Godron (1986) defined landscape as a heterogeneous land area composed of a cluster of interacting ecosystems that is repeated in similar form throughout. The concept differs from the traditional ecosystem concept in focusing on groups of ecosystems and the interactions among them. There are many variants of the definition depending on the research or management context. For example, from a wildlife perspective, we might define landscape as an area of land containing a mosaic of habitat patches, often within which a particular "focal" or "target" habitat patch is embedded (Dunning et al. 1992). Because habitat patches can only be defined relative to a particular organism s perception of the environment (Wiens 1976)(i.e., each organism defines habitat patches differently and at different scales), landscape size would differ among organisms. However, landscapes generally occupy some spatial scale intermediate between an organism s normal home range and its regional distribution. In-other-words, because each organism scales the environment differently (i.e., a salamander and a hawk view their

11 environment on different scales), there is no absolute size for a landscape; from an organism-centered perspective, the size of a landscape varies depending on what constitutes a mosaic of habitat or resource patches meaningful to that particular organism (Fig. 1). 4 Figure 1. Multi-scale view of "landscape" from an organism-centered perspective. Because the eagle, cardinal, and butterfly perceive their environments differently and at different scales, what constitutes a single habitat patch for the eagle may constitute an entire landscape or patch-mosaic for the cardinal, and a single habitat patch for the cardinal may comprise an entire landscape for the butterfly that perceives patches on an even finer scale. This definition most likely contrasts with the more anthropocentric definition that a landscape corresponds to an area of land equal to or larger than, say, a large basin (e.g., several thousand hectares). Indeed, Forman and Godron (1986) suggested a lower limit for landscapes at a "few kilometers in diameter", although they recognized that most of the principles of landscape ecology apply to ecological mosaics at any level of scale. While this may be a more pragmatic definition than the organism-centered definition and perhaps corresponds to our human perception of the environment, it has limited utility in managing

12 5 wildlife populations if you accept the fact that each organism scales the environment differently. From an organism-centered perspective, a landscape could range in absolute scale from an area smaller than a single forest stand (e.g., a individual log) to an entire ecoregion. If you accept this organism-centered definition of a landscape, a logical consequence of this is a mandate to manage wildlife habitats across the full range of spatial scales; each scale, whether it be the stand or watershed, or some other scale, will likely be important for a subset of species, and each species will likely respond to more than 1 scale. KEY POINT It is not our intent to argue for a single definition of landscape. Rather, we wish to point out that there are many appropriate ways to define landscape depending on the phenomenon under consideration. The important point is that a landscape is not necessarily defined by its size; rather, it is defined by an interacting mosaic of patches relevant to the phenomenon under consideration (at any scale). It is incumbent upon the investigator or manager to define landscape in an appropriate manner. The essential first step in any landscape-level research or management endeavor is to define landscape. Patch.--What makes up a landscape? Landscapes are composed of a mosaic of patches (Urban et al. 1987). Landscape ecologists have used a variety of terms to refer to the basic elements or units that make up a landscape, including ecotope, biotope, landscape component, landscape element, landscape unit, landscape cell, geotope, facies, habitat, and site (Forman and Godron 1986). We prefer the term patch, but any of these terms, when defined, are satisfactory according to the preference of the investigator. Like the landscape, patches comprising the landscape are not self-evident; patches must be defined relative to the phenomenon under consideration. For example, from a timber management perspective a patch may correspond to the forest stand. However, the stand may not function as a patch from a particular organism s perspective. From an ecological perspective, patches represent relatively discrete areas (spatial domain) or periods (temporal domain) of relatively homogeneous environmental conditions where the patch boundaries are distinguished by discontinuities in environmental character states from their surroundings of magnitudes that are perceived by or relevant to the organism or ecological phenomenon under consideration (Wiens 1976). From a strictly organism-centered view, patches may be defined as environmental units between which fitness prospects, or "quality", differ; although, in practice, patches may be more appropriately defined by nonrandom distribution of activity or resource utilization among environmental units, as recognized in the concept of "Grain Response" (Wiens 1976). Patches are dynamic and occur on a variety of spatial and temporal scales that, from an organism-centered perspective, vary as a function of each animal s perceptions (Wiens 1976 and 1989a, Wiens and Milne 1989). A patch at any given scale has an

13 internal structure that is a reflection of patchiness at finer scales, and the mosaic containing that patch has a structure that is determined by patchiness at broader scales (Kotliar and Wiens 1990). Thus, regardless of the basis for defining patches, a landscape does not contain a single patch mosaic, but contains a hierarchy of patch mosaics across a range of scales. For example, from an organism-centered perspective, the smallest scale at which an organism perceives and responds to patch structure is its "grain" (Kotliar and Wiens 1990). This lower threshold of heterogeneity is the level of resolution at which the patch size becomes so fine that the individual or species stops responding to it, even though patch structure may actually exist at a finer resolution (Kolasa and Rollo 1991). The lower limit to grain is set by the physiological and perceptual abilities of the organism and therefore varies among species. Similarly, "extent" is the coarsest scale of heterogeneity, or upper threshold of heterogeneity, to which an organism responds (Kotliar and Wiens 1990, Kolasa and Rollo 1991). At the level of the individual, extent is determined by the lifetime home range of the individual (Kotliar and Wiens 1990) and varies among individuals and species. More generally, however, extent varies with the organizational level (e.g., individual, population, metapopulation) under consideration; for example the upper threshold of patchiness for the population would probably greatly exceed that of the individual. Therefore, from an organism-centered perspective, patches can be defined hierarchically in scales ranging between the grain and extent for the individual, deme, population, or range of each species. Patch boundaries are artificially imposed and are in fact meaningful only when referenced to a particular scale (i.e., grain size and extent). For example, even a relatively discrete patch boundary between an aquatic surface (e.g., lake) and terrestrial surface becomes more and more like a continuous gradient as one progresses to a finer and finer resolution. However, most environmental dimensions possess 1 or more "domains of scale" (Wiens 1989a) at which the individual spatial or temporal patches can be treated as functionally homogeneous; at intermediate scales the environmental dimensions appear more as gradients of continuous variation in character states. Thus, as one moves from a finer resolution to coarser resolution, patches may be distinct at some scales (i.e., domains of scale) but not at others. 6 KEY POINT It is not our intent to argue for a particular definition of patch. Rather, we wish to point out the following: (1) that patch must be defined relative to the phenomenon under investigation or management; (2) that, regardless of the phenomenon under consideration (e.g., a species, geomorphological disturbances, etc), patches are dynamic and occur at multiple scales; and (3) that patch boundaries are only meaningful when referenced to a particular scale. It is incumbent upon the investigator or manager to establish the basis for delineating among patches (i.e., patch type classification system) and at a scale appropriate to the phenomenon under consideration.

14 Matrix.--A landscape is composed typically of several types of landscape elements (patches). Of these, the matrix is the most extensive and most connected landscape element type, and therefore plays the dominant role in the functioning of the landscape (Forman and Godron 1986). For example, in a large contiguous area of mature forest embedded with numerous small disturbance patches (e.g., timber harvest patches), the mature forest constitutes the matrix element type because it is greatest in areal extent, is mostly connected, and exerts a dominant influence on the area flora and fauna and ecological processes. In most landscapes, the matrix type is obvious to the investigator or manager. However, in some landscapes, or at a certain point in time during the trajectory of a landscape, the matrix element will not be obvious. Indeed, it may not be appropriate to consider any element as the matrix. Moreover, the designation of a matrix element is largely dependent upon the phenomenon under consideration. For example, in the study of geomorphological processes, the geological substrate may serve to define the matrix and patches; whereas, in the study of vertebrate populations, vegetation structure may serve to define the matrix and patches. In addition, what constitutes the matrix is dependent on the scale of investigation or management. For example, at a particular scale, mature forest may be the matrix with disturbance patches embedded within; whereas, at a coarser scale, agricultural land may be the matrix with mature forest patches embedded within. 7 KEY POINT It is incumbent upon the investigator or manager to determine whether a matrix element exists and should be designated given the scale and phenomenon under consideration. This should be done prior to the analysis of landscape structure since this decision will influence the choice and interpretation of landscape metrics. Scale.--The pattern detected in any ecological mosaic is a function of scale, and the ecological concept of spatial scale encompasses both extent and grain (Forman and Godron 1986, Turner et al. 1989, Wiens 1989). Extent is the overall area encompassed by an investigation or the area included within the landscape boundary. From a statistical perspective, the spatial extent of an investigation is the area defining the population we wish to sample. Grain is the size of the individual units of observation. For example, a fine-grained map might structure information into 1-ha units, whereas a map with an order of magnitude coarser resolution would have information structured into 10-ha units (Turner et al. 1989). Extent and grain define the upper and lower limits of resolution of a study and any inferences about scale-dependency in a system are constrained by the extent and grain of investigation (Wiens 1989). From a statistical perspective, we cannot extrapolate beyond the population sampled, nor can we infer differences among objects smaller than the experimental units. Likewise, in the assessment of landscape structure, we cannot detect pattern beyond the extent of the landscape or below the resolution of the grain (Wiens 1989).

15 As with the concept of landscape and patch, it may be more ecologically meaningful to define scale from the perspective of the organism or ecological phenomenon under consideration. For example, from an organism-centered perspective, grain and extent may be defined as the degree of acuity of a stationary organism with respect to short- and long-range perceptual ability (Kolasa and Rollo 1991). Thus, grain is the finest component of the environment that can be differentiated up close by the organism, and extent is the range at which a relevant object can be distinguished from a fixed vantage point by the organism (Kolasa and Rollo 1991). Unfortunately, while this is ecologically an ideal way to define scale, it is not very pragmatic. Indeed, in practice, extent and grain are often dictated by the scale of the imagery (e.g., aerial photo scale) being used or the technical capabilities of the computing environment. It is critical that extent and grain be defined for a particular study and represent, to the greatest possible degree, the ecological phenomenon or organism under study, otherwise the landscape patterns detected will have little meaning and there is a good chance of reaching erroneous conclusions. For example, it would be meaningless to define grain as 1-ha units if the organism under consideration perceives and responds to habitat patches at a resolution of 1-m 2. A strong landscape pattern at the 1-ha resolution may have no significance to the organism under study. Likewise, it would be unnecessary to define grain as 1-m 2 units if the organism under consideration perceives habitat patches at a resolution of 1-ha. Typically, however, we do not know what the appropriate resolution should be. In this case, it is much safer to choose a finer grain than is believed to be important. Remember, the grain sets the minimum resolution of investigation. Once set, we can always dissolve to a coarser grain. In addition, we can always specify a minimum mapping unit that is coarser than the grain. That is, we can specify the minimum patch size to be represented in a landscape, and this can easily be manipulated above the resolution of the data. It is important to note that the technical capabilities of GIS with respect to image resolution may far exceed the technical capabilities of the remote sensing equipment. Thus, it is possible to generate GIS images at too fine a resolution for the spatial data being represented, resulting in a more complex representation of the landscape than can truly be obtained from the data. Information may be available at a variety of scales and it may be necessary to extrapolate information from one scale to another. In addition, it may be necessary to integrate data represented at different spatial scales. It has been suggested that information can be transferred across scales if both grain and extent are specified (Allen et al. 1987), yet it is unclear how observed landscape patterns vary in response to changes in grain and extent and whether landscape metrics obtained at different scales can be compared. The limited work on this topic suggests that landscape metrics vary in their sensitivity to changes in scale and that qualitative and quantitative changes in measurements across spatial scales will differ depending on how scale is defined (Turner et al. 1989). 8

16 9 Therefore, in investigations of landscape structure, until more is learned, it is critical that any attempts to compare landscapes measured at different scales be done cautiously. KEY POINT One of the most important considerations in any landscape ecological investigation or landscape structural analysis is (1) to explicitly define the scale of the investigation or analysis, (2) to describe any observed patterns or relationships relative to the scale of the investigation, and (3) to be especially cautious when attempting to compare landscapes measured at different scales. Landscape Context.--Landscapes do not exist in isolation. Landscapes are nested within larger landscapes, that are nested within larger landscapes, and so on. In other words, each landscape has a context or regional setting, regardless of scale and how the landscape is defined. The landscape context may constrain processes operating within the landscape. Landscapes are "open" systems; energy, materials, and organisms move into and out of the landscape. This is especially true in practice, where landscapes are often somewhat arbitrarily delineated. That broad-scale processes act to constrain or influence finer-scale phenomena is one of the key principles of hierarchy theory (Allen and Star 1982) and supply-side ecology (Roughgarden et al. 1987). The importance of the landscape context is dependent on the phenomenon of interest, but typically varies as a function of the "openness" of the landscape. The "openness" of the landscape depends not only on the phenomenon under consideration, but on the basis used for delineating the landscape boundary. For example, from a geomorphological or hydrological perspective, the watershed forms a natural landscape, and a landscape defined in this manner might be considered relatively "closed". Of course, energy and materials flow out of this landscape and the landscape context influences the input of energy and materials by affecting climate and so forth, but the system is nevertheless relatively closed. Conversely, from the perspective of a bird population, topographic boundaries may have little ecological relevance, and the landscape defined on the basis of watershed boundaries might be considered a relatively "open" system. Local bird abundance patterns may be produced not only by local processes or events operating within the designated landscape, but also by the dynamics of regional populations or events elsewhere in the species range (Wiens 1981, 1989b, Vaisanen et al. 1986, Haila et al. 1987, Ricklefs 1987). Landscape metrics quantify the structure of the landscape within the designated landscape boundary only. Consequently, the interpretation of these metrics and their ecological significance requires an acute awareness of the landscape context and the openness of the landscape relative to the phenomenon under consideration. These concerns are particularly important for nearest-neighbor metrics. Nearest-neighbor distances are computed solely from patches contained within the landscape boundary. If the landscape

17 extent is small relative to the scale of the organism or ecological processes under consideration and the landscape is an "open" system relative to that organism or process, then nearest-neighbor results can be misleading. Consider a small subpopulation of a species occupying a patch near the boundary of a somewhat arbitrarily defined (from the organism s perspective) landscape. The nearest neighbor within the landscape boundary might be quite far away, yet in reality the closest patch might be very close, but just outside the landscape boundary. The magnitude of this problem is a function of scale. Increasing the size of the landscape relative to the scale at which the organism under investigation perceives and responds to the environment will generally decrease the severity of this problem. In general, the larger the ratio of extent to grain (i.e., the larger the landscape relative to the average patch size), the less likely these and other metrics will be dominated by boundary effects. 10 KEY POINT The important point is that a landscape should be defined relative to both the patch mosaic within the landscape as well as the landscape context. Moreover, consideration should always be given to the landscape context and the openness of the landscape relative to the phenomenon under consideration when choosing and interpreting landscape metrics. Landscape Structure.--Landscapes are distinguished by spatial relationships among component parts. A landscape can be characterized by both its composition and configuration (sometimes referred to as landscape physiognomy or landscape pattern)[dunning et al. 1992, Turner 1989], and these 2 aspects of a landscape can independently or in combination affect ecological processes and organisms. The difference between landscape composition and configuration is analogous to the difference between floristics (e.g., the types of plant species present) and vegetation structure (e.g., foliage height diversity) so commonly considered in wildlife-habitat studies at the within-patch scale. Landscape composition refers to features associated with the presence and amount of each patch type within the landscape, but without being spatially explicit. In other words, landscape composition encompasses the variety and abundance of patch types within a landscape, but not the placement or location of patches within the landscape mosaic. Landscape composition is important to many ecological processes and organisms. For example, many vertebrate species require specific habitat types, and the total amount of suitable habitat (a function of landscape composition) likely influences the occurrence and abundance of these vertebrate species. There have been many attempts to model animal populations within landscapes based on landscape composition alone; such models have been referred to as "island models" by Kareiva (1990). Island models do represent the discrete patchwork mosaic of the landscape; the key feature of these models is population subdivision. Yet these models do not specify the relative distances among

18 patches or their positions relative to each other. Thus, although these models provide strong analytical solutions, they may be overly simplified for most natural populations. It is important to note, however, that we have learned much about population dynamics in spatially complex environments based on models of landscape composition alone (Kareiva 1990). There are many quantitative measures of landscape composition, including the proportion of the landscape in each patch type, patch richness, patch evenness, and patch diversity. Indeed, because of the many ways in which diversity can be measured, there are literally hundreds of possible ways to quantify landscape composition. It is incumbent upon the investigator or manager to choose the formulation that best represents their concerns. Landscape configuration refers to the physical distribution or spatial character of patches within the landscape. Some aspects of configuration, such as patch isolation or patch contagion, are measures of the placement of patch types relative to other patch types, the landscape boundary, or other features of interest. Other aspects of configuration, such as shape and core area, are measures of the spatial character of the patches. There have been many attempts to explicitly incorporate landscape configuration into models of ecological processes and population dynamics within heterogeneous landscapes; such models have been referred to as "stepping-stone models" by Kareiva (1990). In contrast to island models, stepping-stone models have an explicit spatial dimension and can account for dispersal distances and environmental variability with a spatial structure. Recently, there have been dramatic increases in the level of sophistication in stepping-stone models and some results have had profound effects on the design of managed landscapes (e.g., Lamberson et al. 1992, McKelvey et al. 1992). There are many aspects of landscape configuration and the literature is replete with methods and indices developed for representing them. Landscape configuration can be quantified using statistics in terms of the landscape unit itself (i.e., the patch). The spatial pattern being represented is the spatial character of the individual patches. The location of patches relative to each other in the landscape (i.e., the configuration of patches within the landscape), is not explicitly represented. Landscape metrics quantified in terms of the individual patches (e.g., mean patch core area, mean patch shape) are spatially explicit at the level of the individual patch. Such metrics represent a recognition that the ecological properties of a patch are influenced by the surrounding neighborhood (e.g., edge effects) and that the magnitude of these influences are affected by patch size and shape. These metrics simply quantify, for the landscape as a whole, the average patch characteristics or some measure of variability in patch characteristics. Although these metrics are not spatially explicit at the landscape level, they have clear ecological relevance when considered from a patch dynamics standpoint (Pickett and White 1985). For example, a 11

19 number of bird species have been shown to be sensitive to patch core area (a function of patch size and shape) because of negative intrusions from the surrounding landscape (e.g., Temple 1986, Robbins et al. 1989). Quantifying mean patch core area across the landscape could provide a good index to landscape suitability for such species. Landscape metrics quantified in terms of the spatial relationship of patches and matrix comprising the landscape (e.g., nearest neighbor, contagion) are spatially explicit at the landscape level because the relative location of individual patches within the landscape is represented in some way. Such metrics represent a recognition that ecological processes and organisms are affected by the interspersion and juxtaposition of patch types within the landscape. For example, the population dynamics of species with limited dispersal ability are likely affected by the distribution of suitable habitat patches. Both the distance between suitable patches and the spatial arrangement of suitable patches can influence population dynamics (e.g., sensu Kareiva 1990, Lamberson et al. 1992, McKelvey et al. 1992). Likewise, patch juxtaposition is especially important to organisms that require 2 habitat types because the close proximity of resources provided by different patch types is critical for their survival and reproduction. Patch juxtaposition is also important for species adversely affected by edges because the types of patches juxtaposed along an edge will influence the character of that edge. A number of landscape configuration metrics can be formulated either in terms of the individual patches or in terms of the whole landscape, depending on the emphasis sought. For example, fractal dimension is a measure of shape complexity (Mandelbrot 1982, Burrough 1986, Milne 1988) that can be computed for each patch and then averaged for the landscape, or it can be computed from the landscape as a whole (e.g., using the box-count method, Morse et al. 1985). Similarly, core area can be computed for each patch and then represented as mean patch core area for the landscape, or it can be computed simply as total core area in the landscape. Obviously, one form can be derived from the other if the number of patches is known and so they are largely redundant; the choice of formulations is dependent upon user preference or the emphasis (patch or landscape) sought. The same is true for a number of other common landscape metrics. Typically, these metrics are spatially explicit at the patch level but not at the landscape level. Not all landscape metrics can easily be classified as representing landscape composition or landscape configuration. For example, landscape metrics such as mean patch size and patch density are not really spatially explicit at either the patch or landscape level because they do not depend explicitly on the spatial character of the patches or their relative location. Moreover, mean patch size and patch density of a particular patch type reflect both the amount of a patch type present (composition) and its spatial distribution (configuration). Because mean patch size and patch density vary as a function of the spatial pattern complexity of the landscape, it is often more appropriate to consider these 12

20 indices of landscape configuration. In addition, there are some landscape metrics that clearly represent pattern complexity but are not spatially explicit at all. These metrics vary as a function of the heterogeneity of the landscape, but do not depend explicitly on the relative location of patches within the landscape or their individual spatial character. For example, total edge or edge density is a function of the amount of border between patches. For a given edge density there could be 2 patches or 10 patches, they could be clustered or maximally dispersed, or they could be skewed to one side of the landscape or in the middle. It is not important that all metrics be classified according to the simple composition versus configuration dichotomy. What is important, however, is that the investigator or manager recognize that landscape structure consists of both composition and configuration and that various metrics have been developed to represent these aspects of landscape structure separately or in combination. Finally, it is important to understand how measures of landscape structure are influenced by the designation of a matrix element. If an element is designated as matrix and therefore presumed to function as such (i.e., has a dominant influence on landscape dynamics), then it should not be included as another "patch" type in any metric that simply averages some characteristic across all patches (e.g., mean patch size, mean patch shape). Otherwise, the matrix will dominate the metric and serve more to characterize the matrix than the patches within the landscape, although this may itself be meaningful in some applications. From a practical standpoint, it is important to recognize this because in FRAGSTATS the matrix can be excluded from calculations by designating its class value as background. If the matrix is not excluded from the calculations, it may be more meaningful to use the class-level statistics for each patch type and simply ignore the patch type designated as the matrix. From a conceptual standpoint, it is important to recognize that the choice and interpretation of landscape metrics must ultimately be evaluated in terms of their ecological meaningfulness, which is dependent upon how the landscape is defined, including the choice of patch types and the designation of a matrix. 13 KEY POINT The importance of fully understanding each landscape metric before it is used cannot be emphasized enough. Specifically, these questions should be asked of each metric before it is used: does it represent landscape composition, configuration, or both; what aspect of configuration does it represent; what scale, if any, is spatially explicit; how is it affected by the designation of a matrix element? Based on answers to these questions, does the metric represent landscape structure in a manner ecologically meaningful to the phenomenon under consideration? Only after answering these questions should one attempt to draw conclusions about the structure of the landscape analyzed.

21 14 FRAGSTATS OVERVIEW FRAGSTATS is a spatial pattern analysis program for quantifying landscape structure. The landscape subject to analysis is user-defined and can represent any spatial phenomenon. FRAGSTATS quantifies the areal extent and spatial distribution of patches (i.e., polygons on a map coverage) within a landscape; it is incumbent upon the user to establish a sound basis for defining and scaling the landscape (including the extent and grain of the landscape) and the scheme upon which patches within the landscape are classified and delineated (we strongly recommend that you read the preceding section on Concepts and Definitions). The output from FRAGSTATS is meaningful only if the landscape mosaic is meaningful relative to the phenomenon under consideration. FRAGSTATS does not limit the scale (extent or grain) of the landscape subject to analysis. However, the distance- and area-based metrics computed in FRAGSTATS are reported in meters and hectares, respectively. Thus, landscapes of extreme extent and/or resolution may result in rather cumbersome numbers and/or be subject to rounding errors. However, FRAGSTATS outputs data files in ASCII format that can be manipulated using any database management program to rescale metrics or to convert them to other units (e.g., converting hectares to acres). There are 2 versions of FRAGSTATS; one that accepts Arc/Info polygon coverages (vector), and one that accepts a raster image in a variety of formats. The vector version of FRAGSTATS is an Arc/Info AML developed on a SUN workstation using Arc/Info version 6.1; it will not run with earlier versions of Arc/Info. Because of limitations in Arc/Info, the AML calls several C programs that were developed in a Unix environment and compiled with the GNU C compiler (note, they may not compile with other compilers). The raster version of FRAGSTATS also was developed on a SUN workstation in the Unix operating environment. It is written in C and also compiled with the GNU C compiler. Both versions of FRAGSTATS respond to command line input or allow the user to answer a series of prompts. Both versions of FRAGSTATS generate the same array of metrics (see Table 1), although a few additional metrics (i.e., nearest-neighbor metrics and contagion) are computed in the raster version, and the format of the output files is exactly the same. The raster version of FRAGSTATS also has been compiled to run in the DOS environment on a personal computer (PC). The directions for running the DOS version on a PC are exactly the same as the Unix version. It is important to realize that vector and raster images depict edges differently. Vector images portray a line in the form it is digitized. Raster images, however, portray lines in stair-step fashion. Consequently, the measurement of edge length is biased upward in raster images; that is, measured edge length is always more than the true edge length. The magnitude of this bias depends on the grain or resolution of the image (i.e., cell size),